cloneofsimo / karras-power-ema-tutorial
☆51Updated last year
Alternatives and similar repositories for karras-power-ema-tutorial:
Users that are interested in karras-power-ema-tutorial are comparing it to the libraries listed below
- Simple implementation of muP, based on Spectral Condition for Feature Learning. The implementation is SGD only, dont use it for Adam☆73Updated 7 months ago
- These papers will provide unique insightful concepts that will broaden your perspective on neural networks and deep learning☆47Updated last year
- ☆26Updated 10 months ago
- ☆33Updated 5 months ago
- A general framework for inference-time scaling and steering of diffusion models with arbitrary rewards.☆92Updated 2 weeks ago
- ☆32Updated 4 months ago
- WIP☆93Updated 6 months ago
- ☆37Updated 10 months ago
- Focused on fast experimentation and simplicity☆68Updated 2 months ago
- Official codebase for Margin-aware Preference Optimization for Aligning Diffusion Models without Reference (MaPO).☆68Updated 8 months ago
- The codebase of our paper "Improving the Training of Rectified Flows", NeurIPS 2024☆97Updated 4 months ago
- ☆31Updated last year
- ☆21Updated 8 months ago
- ☆13Updated 9 months ago
- [ICML 2023] Reflected Diffusion Models (https://arxiv.org/abs/2304.04740)☆157Updated last year
- JAX implementation ViT-VQGAN☆82Updated 2 years ago
- Train VAE like a boss☆265Updated 4 months ago
- ☆33Updated 5 months ago
- ☆25Updated 9 months ago
- CLOOB training (JAX) and inference (JAX and PyTorch)☆70Updated 2 years ago
- Official code for "On Calibrating Diffusion Probabilistic Models"☆29Updated 2 years ago
- Official code implementation for the work Preference Alignment with Flow Matching (NeurIPS 2024)☆41Updated 3 months ago
- Minimal multi-gpu implementation of EDM2: "Analyzing and Improving the Training Dynamics of Diffusion Models"☆28Updated 11 months ago
- The official implementation of Diffusion-KTO: Aligning Diffusion Models by Optimizing Human Utility☆37Updated last month
- ☆75Updated 7 months ago
- Implementation of "Analyzing and Improving the Training Dynamics of Diffusion Models"☆92Updated last year
- Official implementation for Rare-to-Frequent (R2F), ICLR'25, Spotlight☆37Updated 2 weeks ago
- ☆27Updated last month
- Minimal (400 LOC) implementation Maximum (multi-node, FSDP) GPT training☆122Updated 10 months ago